1. Field of the Invention:
This invention relates to a data processing device and more particularly to a device for processing data, which include a first data sequence indicative of the amplitude of an information signal and a second data sequence each data of which is indicative of the error rate of each data of the first data sequence.
2. Description of the Prior Art:
Digital signal reproducing apparatuses in general are sometimes prevented from accurately reproducing data by occurrences of noises produced in the data transmission line, a time crosstalk, a drop-out, etc. Generally, such noise is produced in a pulse like shape and is sometimes reproduced in an excessive degree depending on the part at which a data error is produced by the noise within a data sequence. In view of this, error correcting or error detecting codes, etc., are added to a data sequence to impart redundancy to the data sequence for the purpose of correcting an error in the recorded information by real time processing. Then, for correct data reproduction, preceding or previous value holding, intermediate value interpolation, etc., are sometimes carried out on the side of a reproducing apparatus to compensate for errors due to noises. However, there might take place such a large data drop-out that is unsuppressible by the above-stated error compensation arrangement heretofore employed against noises. In that event, a data which is produced in a sequential, time serial manner is quite likely to become erroneous. For example, in the event of such a large drop-out in the data of a digital signal obtained by pulse code modulation (PCM) of an audio signal, there will be produced an abnormal sound. Therefore, to solve this problem, it has been practiced to use, as necessary, the above-stated error detecting codes, error correcting codes, etc. for forming flag data which indicate the error rate of each data; and the output signal is muted by changing the output data into data corresponding to a zero level when the flag data come to consecutively show a high error rate over a long period.
However, in muting the output signal, if the output data is immediately changed to a data corresponding to the zero level, the muting action brings about a sudden change in amplitude. This is undesirable especially for an audio signal, in terms of auditory sensation. To prevent this, there is a method, in which the above-stated flag data are arranged to be counted for a predetermined period of time and to be attenuated by a plurality of attenuators according to a count value thus obtained, so that the muting action can be gradually performed. This method eliminates sudden amplitude changes of the output signal as muting is arranged to be accomplished by gradually attenuating the data and to be gradually undone. This method, however, requires extremely complex circuit arrangement and thus has been hardly applicable to houshold appliances which must be compact and not expensive.